Cortical morphological alterations in vestibular migraine: insights from surface-based morphometry and machine learning.

BACKGROUND: Previous surface-based morphometry (SBM) research on cortical morphology in vestibular migraine (VM) has been limited by a small sample size. This study aims to validate cortical morphological alterations in VM using SBM and explore their clinical implications with a larger sample. METHODS: Fifty-five patients with VM and 65 healthy controls (HCs) underwent structural T1-weighted MRI. Cortical morphological features, including cortical thickness, curvature, surface area, and local gyrification index, were assessed using SBM with FreeSurfer. Statistical analyses were conducted to examine inter-group differences and correlations between cortical alterations and clinical features. A linear support vector machine (SVM) classifier was applied to evaluate the performance of cortical morphological differences in distinguishing VM patients from HCs. RESULTS: The SBM analysis revealed significant cortical differences between the VM group and HCs. Specifically, cortical thickness was significantly reduced in the right superior frontal gyrus, superior parietal lobule, and precuneus in the VM group as compared to HCs. Additionally, surface area was significantly smaller in the right rostral middle frontal cortex in the VM cohort. No significant correlations were found between the cortical morphological features and any clinical indicators. The SVM classification model achieved moderate efficacy (area under the curve = 0.775, p < 0.001) in distinguishing VM patients from HCs. CONCLUSIONS: These findings demonstrate significant cortical morphological alterations in the frontoparietal regions of patients with VM, which may be associated with dizziness, pain, and emotional and cognitive dysfunctions. The identified cortical differences have the potential to serve as neuroimaging markers for VM.